CN-121998765-A - Intelligent commodity purchase and sale matching and risk hedging method and system

CN121998765ACN 121998765 ACN121998765 ACN 121998765ACN-121998765-A

Abstract

The invention discloses an intelligent purchase and sale matching and risk hedging method and system for commodities, which belong to the technical field of data processing, and are characterized in that a low-frequency risk factor sequence is generated by receiving multisource supply chain data and high-frequency financial market quotation data through sliding window filtering, a fusion data set is formed by time alignment of the market quotation data and the supply chain data, a supply and demand bipartite graph is built based on the fusion data set, logistics, fund cost and risk cost quantized by the low-frequency risk factor are comprehensively considered, a global optimal matching scheme meeting the delivery constraint is solved, purchase and sale matching orders are generated, a dynamic risk assessment model is further built, a multi-scenario price path is generated by combining order terms and the low-frequency risk factor through Monte Carlo simulation, potential damage distribution is output, a bound derivative hedging proposal is generated according to the potential damage distribution, and finally an integrated transaction hedging scheme is formed. The invention realizes the collaborative optimization of the supply chain performance and the financial risk, and obviously improves the matching efficiency and the risk resistance.

Inventors

YANG XINYONG
XIE WENBIN
LUO JING

Assignees

广州广美科创有限公司

Dates

Publication Date: 20260508
Application Date: 20260127

Claims (10)

1. The intelligent commodity purchase and sale matching and risk hedging method is characterized by comprising the following steps of: receiving multi-source supply chain data and financial market quotation data; Performing sliding window filtering processing on the financial market quotation data to generate a low-frequency risk factor sequence, and performing time alignment on the low-frequency risk factor sequence and the multi-source supply chain data to construct a fusion data set; Constructing a supply and demand bipartite graph based on the fusion data set and calculating comprehensive weights of edges, wherein the comprehensive weights comprise cost items and risk cost items determined based on the low-frequency risk factor sequences; under the constraint conditions of meeting the delivery time, the delivery place, the commodity specification and the quantity, solving a matching scheme which enables the global comprehensive weight to be optimal, and generating purchase and sale matching order data; Building a dynamic risk assessment model, wherein the dynamic risk assessment model generates price paths under a plurality of market scenes through Monte Carlo simulation based on purchase and sales matching order data and the low-frequency risk factor sequence, calculates cash flow aggregation output potential damage distribution data under a plurality of scenes by combining with order delivery terms, and generates derivative hedging advice bound with the purchase and sales matching order data based on the output potential damage distribution data; And generating an integrated transaction risk avoidance scheme containing the purchase and sales matching order data and the hedging advice.
2. The intelligent purchase-sale matching and risk hedging method for commodities according to claim 1, wherein, The fusion data set comprises a plurality of purchasing demand nodes and a plurality of supply offer nodes, wherein the purchasing demand nodes comprise demand commodities, demand quantity, expected delivery places and time window information, and the supply offer nodes comprise supply commodities, supply quantity, deliverable places and time window information; Calculating an expected logistics cost based on the deliverable location in the supply offer node and the expected delivery location in the procurement demand node in combination with logistics data in the multi-source supply chain data; calculating an expected funds occupation cost based on the supply price in the supply offer node, the demand in the purchase demand node and a preset funds utilization rate; Based on an analysis of the corresponding sequence of commodity prices in the current financial market quotation data, an expected price volatility risk cost during expected order fulfillment is calculated as the expected risk cost.
3. The intelligent purchase-sale matching and risk hedging method of a commodity according to claim 2, further comprising: Taking all feasible supply and demand node connection relations in the bipartite graph as decision variables, wherein the values of the decision variables represent whether a corresponding pair of supply and demand nodes are matched or not; Minimizing the sum of the comprehensive weights corresponding to all the matched edges as an optimization target; Taking the constraint conditions that the total matching amount of each purchasing demand node does not exceed the demand amount of the purchasing demand node, the total matching amount of each supply offer node does not exceed the supply amount of the purchasing demand node, and the commodity specification, the delivery time and the delivery place meet the preset tolerance range; And solving by adopting a mixed integer linear programming solver based on the decision variables, the optimization targets and the constraint conditions to obtain a matching result.
4. The intelligent purchase-sale matching and risk hedging method for commodities according to claim 1, wherein the constructing of the dynamic risk assessment model specifically comprises: Dynamically adjusting a drift term of a random process model for describing commodity price evolution based on the delivery time window of purchase and sales matching order data so as to reflect expected changes of the market in a delivery period; Extracting a spot price sequence and a futures price sequence corresponding to commodity targets related to the purchase and sales matching order data from the financial market quotation data, and carrying out parameter calibration on the random process model based on the sequences; Taking the calibrated random process model, the delivery terms of purchase and sale matching order data and credit state information of a transaction counter party as inputs, and adopting a Monte Carlo simulation method based on importance sampling to only perform key sampling and simulation on a tail risk path which possibly causes significant loss, rather than uniformly sampling a full-scale price path; The importance sampling improves the path generation probability under the extreme market situation under the guidance of the adjusted drift term, and reduces the calculation complexity on the premise of determining the risk assessment accuracy.
5. The intelligent purchase-sale matching and risk hedging method for commodities according to claim 4, wherein the working process of outputting the potential damage distribution data comprises: Establishing an importance sampling optimized calibrated random process model, and generating a group of simulated price paths focused on a tail risk area, wherein each path covers a time interval from the current moment to the final delivery date of an order; Calculating corresponding expected cash flows at each key performance time point on each simulated price path according to the delivery terms of the purchase and sales matching order data; Applying credit risk adjustment to the expected cash flow by combining credit status information of the transaction counter party to obtain an adjusted cash flow; And according to the adjusted cash flow obtained under the key sampling path, mapping and statistics aggregation are carried out, potential damage distribution data of the purchase and sales matching order data is generated, and unbiased correction is carried out on the distribution based on sampling weight, so that the statistical consistency of the evaluation result is ensured.
6. The intelligent purchase-sale matching and risk hedging method of merchandise of claim 1, wherein generating derivative hedging advice bound to the purchase-sale matching order data further comprises: Carrying out statistical analysis on the potential damage distribution data, and extracting risk measurement indexes of the potential damage distribution data under a preset confidence level; determining the type, expiration period and row weight interval of the target hedging tool according to the risk measurement index; Selecting a plurality of standardized derivative contracts in the expiration period and the row weight interval to form an initial hedging combination; And optimizing and calculating positions of all contracts in the initial hedging combination by taking a risk measurement index of the minimized hedging combination and meeting a preset maximum hedging cost constraint as a target to obtain a position configuration scheme, wherein the position configuration scheme forms the derivative hedging proposal.
7. The intelligent purchase-sale matching and risk hedging method of a commodity according to claim 1 or 6, further comprising: Receiving a confirmation instruction of a user to the integrated transaction risk avoidance scheme; generating structured electronic contract data according to the terms of the purchase and sales matching order data and triggering an electronic signing process in response to the confirmation instruction; When the potential damage distribution data output by the dynamic risk assessment model exceeds a preset threshold, automatically triggering recalculation of the purchase and sales matching model, generating an alternative matching scheme, and locking a corresponding financial transaction system interface until the hedging instruction is confirmed.
8. The intelligent purchase-sale matching and risk hedging method of commodity according to claim 1, wherein said multi-source supply chain data further comprises credit data of a transaction partner, said credit data comprising financial information disclosed by an enterprise and historical transaction performance information.
9. An intelligent commodity purchase-sale matching and risk hedging system, which applies the intelligent commodity purchase-sale matching and risk hedging method as claimed in claims 1-8, and is characterized by comprising the following steps: The data aggregation module is used for aggregating multi-source supply chain data and financial market quotation data in real time, wherein the multi-source supply chain data comprises capacity data, inventory data and logistics data, and the financial market quotation data comprises commodity spot quotation data and futures quotation data; The intelligent matching module is used for generating purchase and sale matching order data for a purchasing party and a supplier based on the aggregated multi-source supply chain data and the financial market quotation data, and comprises: the supply and demand modeling unit is used for modeling a plurality of purchasing demands and a plurality of supply offers into two-side nodes of the bipartite graph respectively; the weight calculation unit is used for calculating comprehensive weights of edges connecting the pair of the nodes aiming at supply and demand nodes matched with each pair of commodity specifications, wherein the comprehensive weights comprise commodity basic price cost, expected logistics cost, expected capital occupation cost and expected risk cost obtained based on preliminary risk analysis on the current financial market quotation data; The optimization solving unit is used for constructing and solving a multi-objective optimization model with constraint based on the bipartite graph, the comprehensive weight and a preset hard constraint condition so as to obtain an optimal matching scheme; The order generation unit is used for generating the purchase and sale matching order data according to the optimal matching scheme; The risk simulation module is used for constructing a dynamic risk assessment model aiming at the purchase and sales matching order data so as to output potential damage distribution data of the purchase and sales matching order data under preset various market fluctuation situations; The hedging suggestion generation module is used for generating derivative hedging suggestions intelligently bound with the purchase and sales matching order data based on the potential damage distribution data; and the scheme output module is used for generating an integrated transaction risk avoiding scheme by the purchase and sale matching order data and the derivative hedging proposal and outputting the integrated transaction risk avoiding scheme to a user.
10. A computer readable storage medium having stored thereon a computer program, wherein the computer program when executed by a processor performs the steps of the intelligent purchase-and-sale matching and risk hedging method of a commodity according to any one of claims 1 to 8.

Description

Intelligent commodity purchase and sale matching and risk hedging method and system Technical Field The invention belongs to the technical field of data processing, and particularly relates to an intelligent commodity purchase and sale matching and risk hedging method and system. Background In the commodity trade field, there is the problem that information asymmetry, supply and demand matching inefficiency for a long time exist between purchasing side and the supplier. Traditional purchase and sale matching relies on manual negotiation or simple information release of an information platform, the decision process is slow, and optimal allocation of resources is difficult to realize in a dynamic market environment. Meanwhile, commodity price fluctuation is severe, and huge price risk exposure is brought to both parties of the transaction. While there are separate inventory warning systems or platforms for providing purchase and sales information interfacing, and theory and practice of risk hedging using financial derivatives, in the prior art, these solutions tend to be fraught. The transaction matching system generally does not provide real-time risk assessment which is bound with specific transaction depth, but the use of a risk management tool (such as futures and options) requires professional knowledge of a transaction party, and complex calculation is carried out independently of a transaction decision, so that the problems of high operation threshold, strong hysteresis and difficulty in accurately matching specific transaction risks exist. Therefore, how to deeply integrate the intelligent matching of the entity transaction with the real-time management of the financial risk in the algorithm level, and synchronously generate a customized risk avoiding scheme while achieving the transaction becomes a key bottleneck for improving the commodity trade efficiency and robustness. Disclosure of Invention The invention aims to provide an intelligent commodity purchase and sale matching and risk hedging method and system, which are used for realizing intelligent optimization matching of purchase and sale orders, multi-scenario dynamic risk simulation and automatic generation and binding of customized derivative hedging suggestions by constructing a real-time perception network and an intelligent analysis model containing multi-source supply chain and financial market data, and realizing deep fusion of traditional isolated transaction matching and risk management on an algorithm level to form an integrated transaction hedging scheme, so that the technical problems of low purchase and sale matching efficiency, high price fluctuation risk and disjoint management of the purchase and sale matching efficiency and the price fluctuation risk in commodity transaction are effectively solved, the decision intelligentization level, the risk management and control accuracy and the operation robustness of commodity trade are obviously improved, and technical support is provided for digital and intelligent conversion of supply chain finance. In order to achieve the above object, in a first aspect of the present invention, there is provided a method for intelligent purchase-sale matching and risk hedging of a commodity, comprising the steps of: receiving multi-source supply chain data and financial market quotation data; Performing sliding window filtering processing on the financial market quotation data to generate a low-frequency risk factor sequence, and performing time alignment on the low-frequency risk factor sequence and the multi-source supply chain data to construct a fusion data set; Constructing a supply and demand bipartite graph based on the fusion data set and calculating comprehensive weights of edges, wherein the comprehensive weights comprise cost items and risk cost items determined based on the low-frequency risk factor sequences; under the constraint conditions of meeting the delivery time, the delivery place, the commodity specification and the quantity, solving a matching scheme which enables the global comprehensive weight to be optimal, and generating purchase and sale matching order data; Building a dynamic risk assessment model, wherein the dynamic risk assessment model generates price paths under a plurality of market scenes through Monte Carlo simulation based on purchase and sales matching order data and the low-frequency risk factor sequence, calculates cash flow aggregation output potential damage distribution data under a plurality of scenes by combining with order delivery terms, and generates derivative hedging advice bound with the purchase and sales matching order data based on the output potential damage distribution data; And generating an integrated transaction risk avoidance scheme containing the purchase and sales matching order data and the hedging advice. Wherein the financial market quotation data is high frequency financial market quotation data. Further, the fusion data set compr